JP4708250B2 - Red-eye correction processing system, red-eye correction processing method, and red-eye correction processing program - Google Patents

Description

  The present invention relates to a red-eye correction processing system, a red-eye correction processing method, and a red-eye correction processing program. In particular, the present invention relates to a red-eye correction processing system, a red-eye correction processing method, and a red-eye correction processing program for creating an index print.

  When a strobe is used at the time of imaging, a red-eye phenomenon in which a human pupil appears in red or a gold-eye phenomenon in which gold appears in gold may occur. When the red-eye phenomenon or the gold-eye phenomenon occurs, it cannot be said that the photographed subject looks good. Therefore, various image processing methods for correcting these red-eye or gold-eye to natural pupil colors have been proposed. For example, the user is allowed to specify a processing target area including an image of an eye with poor color tone, and based on the color value obtained from the hue, saturation, and brightness of the specified processing target area, A method of performing red-eye correction processing has been proposed (see, for example, Patent Document 1). In addition, shooting information such as strobe information, Ev value, shutter speed, and aperture value is added to the captured image, and red-eye correction processing is performed only on an image that is determined to have a possible red eye. A method has been proposed (see, for example, Patent Document 4). In order to identify the contents of image processing performed on a plurality of images, a system for creating an index print in which characters indicating the contents of image processing are added to images has been proposed (for example, Patent Document 6). reference.).

In addition, after extracting a candidate area including a human face from an image, the extracted candidate area is divided into a plurality of small areas, and the feature amounts of the plurality of small areas are collated with a preset face area pattern. A method for extracting a face area (see, for example, Patent Document 2), a human face area extracted from an image, and a human body candidate when the color density of the extracted face area corresponds to a predetermined threshold The face area is extracted by evaluating the accuracy of the face area (the likelihood that the extracted area is a face image) using the density and saturation of the face or torso area. And a method of extracting a plurality of face candidate regions from an image and evaluating the accuracy from the degree of overlap of the extracted plurality of face candidate regions to extract a face region. (For example, see Patent Document 5) ).
JP 2000-76427 A JP 2000-137788 A JP 2000-148980 A JP 2004-145287 A JP 2000-149018 A Japanese Patent Laid-Open No. 11-341256

  However, in the index print output in the invention described in Patent Document 5, for example, when the image processing performed on the image is a red-eye correction processing, the type of red-eye (red-eye, gold-eye, and Regardless of the type (such as black eye), only the information indicating whether or not the red-eye correction process has been applied to the image is output. Therefore, even if the user refers to the index print, the user may not be able to determine what eye the red-eye correction target eye included in the original image was.

  Accordingly, an object of the present invention is to provide a red-eye correction processing system, a red-eye correction processing method, and a red-eye correction processing program that can solve the above-described problems. This object is achieved by a combination of features described in the independent claims. The dependent claims define further advantageous specific examples of the present invention.

  In order to solve the above problems, in the first embodiment of the present invention, a red-eye correction processing system includes an image input unit that inputs a plurality of images and a plurality of images that are input by the image input unit. The red-eye detection unit for detecting the red eye, the red-eye correction unit for performing the red-eye correction process detected by the red-eye detection unit, and whether the correction degree of the red-eye correction process by the red-eye correction unit is greater than a predetermined correction level A correction degree determination unit that determines whether or not a red-eye is not detected by the red-eye detection unit, an image that has not been subjected to red-eye correction processing is output, and a red-eye that is detected by the red-eye detection unit Of the images output by the image output unit that outputs the corrected image and the images output by the image output unit, red-eye correction processing is performed on an image that has been subjected to red-eye correction processing that is greater than a predetermined correction degree. Information indicating that the red-eye correction processing has been performed is not output in association with the image. A correction result output unit.

  Further, the image processing apparatus further includes a red-eye color information acquisition unit that acquires pre-correction color information that is information indicating the red-eye color before red-eye correction detected by the red-eye detection unit, and the correction degree determination unit is acquired by the red-eye color information acquisition unit Based on the pre-correction color information, it may be determined whether the correction degree of the red-eye correction process by the red-eye correction unit is greater than a predetermined correction degree. The red-eye color information acquisition unit further acquires post-correction color information that is information indicating the color of the red eye after the red-eye correction unit has performed the red-eye correction process, and the correction degree determination unit includes a red-eye color information acquisition unit When the pre-correction color information acquired and the post-correction color information are compared, and the amount of change from the color indicated by the pre-correction color information of the color indicated by the post-correction color information is included within a predetermined range, You may determine that the correction degree by the red eye correction process of the red eye correction unit is smaller than a predetermined correction degree.

  The red-eye color information acquisition unit acquires information indicating the color balance in the red-eye region before red-eye correction detected by the red-eye detection unit as pre-correction color information, and the correction degree determination unit indicates the pre-correction color information. When the color balance is a predetermined balance, it may be determined that the correction degree by the red-eye correction process of the red-eye correction unit is smaller than the predetermined correction degree. The red-eye detection unit further acquires red-eye region information indicating the size of the detected red-eye region, and the correction degree determination unit is based on the red-eye region information acquired by the red-eye detection unit and corrects the red-eye by the red-eye correction unit. It may be determined whether or not the processing correction degree is greater than a predetermined correction degree. Further, the red-eye detection unit further acquires red-eye position information indicating a position in the detected red-eye image, and the correction degree determination unit is based on the red-eye position information acquired by the red-eye detection unit, and the red-eye correction processing by the red-eye correction unit It may be determined whether the correction degree is greater than a predetermined correction degree.

  The image input unit further includes a black eye detection unit that detects black eyes included in each of the plurality of images input, and the correction degree determination unit includes the black eye detection unit at the same position as the red eye detected by the red eye detection unit. May be determined that the correction degree by the red-eye correction process of the red-eye correction unit is smaller than a predetermined correction degree. Furthermore, the red-eye detection unit matches a red-eye candidate and a red-eye template image included in each of the plurality of images, detects a red-eye candidate having a predetermined first matching degree or more as a red-eye, and a correction degree determination unit When the red-eye correction unit performs red-eye correction processing with a red eye smaller than a predetermined second matching level that is larger than the first matching degree, the correction degree of the red-eye correction processing by the red-eye correction unit is a predetermined correction degree. You may judge that it is smaller than.

  The image input unit further includes an imaging condition determination unit that determines whether or not the image input when the strobe is fired. The correction degree determination unit emits the strobe light by the imaging condition determination unit. It may be determined that the correction degree of the image that is determined to have been captured when the image is not captured is smaller than a predetermined correction degree. The imaging condition determining unit determines whether or not the luminance of the subject included in the image input by the image input unit is equal to or higher than a predetermined value. It may be determined that the correction degree of the image for which the luminance of the image is determined to be equal to or higher than a predetermined value is smaller than the predetermined correction degree. The imaging condition determination unit determines whether or not the distance from the imaging device to the subject at the time of imaging of the image input by the image input unit is within a predetermined range. The degree of correction of an image for which the distance is determined not to be within a predetermined range by the determination unit may be determined to be smaller than a predetermined correction degree.

  The imaging condition determination unit determines whether the image is captured indoors from light source information at the time of imaging of the image input by the image input unit, and the correction degree determination unit determines whether the image is captured indoors by the imaging condition determination unit. You may determine that the correction degree about the image judged to be imaged is larger than a predetermined correction degree. Further, the imaging condition determination unit determines whether the imaging mode at the time of imaging of the image input by the image input unit is the portrait mode or the night view mode, and the correction degree determination unit determines whether the image is captured by the imaging condition determination unit. You may determine that the correction degree about the image determined to be imaged in portrait mode or night view mode is larger than a predetermined correction degree. The imaging condition determination unit determines whether the image input by the image input unit is subjected to red-eye correction processing, and the correction degree determination unit performs the red-eye correction processing on the image by the imaging condition determination unit. It may be determined that the correction degree of the image determined to be smaller than a predetermined correction degree.

  Further, in the second embodiment of the present invention, a red-eye correction processing method for performing red-eye correction processing on an image, each of an image input stage for inputting a plurality of images and a plurality of images input in the image input stage A red-eye detection stage for detecting red eyes, a red-eye correction stage for performing red-eye correction processing detected in the red-eye detection stage, and a predetermined correction degree for which the correction degree of the red-eye correction processing by the red-eye correction stage is predetermined. An image that has not been subjected to red-eye correction processing is output for an image in which the red-eye is not detected in the correction degree determination stage for determining whether or not the red-eye is detected in the red-eye detection stage, and an image in which red-eye is detected in the red-eye detection stage The image output stage for outputting the image subjected to the red-eye correction process, and the image output in the image output stage is larger than a predetermined correction degree. Information indicating that red-eye correction processing has been performed is output for an image subjected to eye correction processing in association with the image, and red-eye correction processing is performed on an image that has undergone red-eye correction processing that is less than a predetermined correction degree. A correction result output stage that does not output information indicating that it has been applied in association with an image.

  According to a third aspect of the present invention, there is provided a red-eye correction processing program for a red-eye correction processing system that performs red-eye correction processing on an image. The red-eye correction processing system includes an image input unit that inputs a plurality of images, an image A red-eye detection unit that detects red-eye included in each of a plurality of images input by the input unit, a red-eye correction unit that performs red-eye correction processing detected by the red-eye detection unit, and a correction degree of red-eye correction processing by the red-eye correction unit in advance A correction degree determination unit that determines whether or not a predetermined correction degree is greater than a predetermined correction degree, and outputs an image that has not been subjected to red-eye correction processing for an image that has not been detected by the red-eye detection unit. For an image in which red-eye is detected, an image output unit that outputs an image subjected to red-eye correction processing, and an image that is output from the image output unit has a red value that is greater than a predetermined correction level. Information indicating that red-eye correction processing has been performed is output to the image subjected to correction processing in association with the image, and red-eye correction processing is performed on an image that has been subjected to red-eye correction processing that is less than a predetermined correction degree. It is made to function as a correction result output part which does not output the information which shows having been made corresponding to an image.

  The above summary of the invention does not enumerate all the necessary features of the present invention, and sub-combinations of these feature groups can also be the invention.

  According to the present invention, it is possible to determine whether to display a red-eye correction result according to the degree of red-eye correction processing, or to add eye information of the original image to the red-eye correction result.

  Hereinafter, the present invention will be described through embodiments of the invention. However, the following embodiments do not limit the invention according to the scope of claims, and all combinations of features described in the embodiments are included. It is not necessarily essential for the solution of the invention.

  FIG. 1 shows an overview of a red-eye correction processing system 10 according to an embodiment of the present invention. The accepting device 12, the image processing device 14, the printer 16, and the personal computer 18 are examples of the red-eye correction processing system 10. When outputting an index print, the red-eye correction processing system 10 according to the present embodiment outputs information indicating that red-eye correction processing has been performed when the degree of red-eye correction processing performed on an image is equal to or greater than a predetermined level. On the other hand, when the degree of the red-eye correction process performed on the image is equal to or less than the predetermined degree, the object is to not output information indicating that the red-eye correction process has been performed. In addition, the red-eye correction processing system 10 according to the present embodiment is intended to output information indicating what kind of eye the red-eye correction target included in the original image is when an index print is output. .

  The red-eye correction processing system 10 according to the present embodiment receives image data from a user. Then, the red-eye correction processing system 10 performs red-eye correction processing on the image indicated by the image data received from the user. Here, the red eye may include a gold eye in addition to the red eye. Note that the red-eye correction processing system 10 does not perform the red-eye correction process on the image when the image does not include an eye to be subjected to the red-eye correction process. Subsequently, the red-eye correction processing system 10 acquires the information about the eyes included in the input image and the information about the eyes included in the image after the red-eye correction processing, and compares the acquired information. . Then, the red-eye correction processing system 10 determines the correction degree of the red-eye correction process based on the comparison result. When the red-eye correction processing system 10 outputs an image subjected to red-eye correction processing, the red-eye correction processing system 10 adds a mark or the like according to the correction degree of the red-eye correction processing to the image.

  For example, the red-eye correction processing system 10 compares the information related to the eyes included in the input image with the information related to the eyes included in the image after the red-eye correction processing is performed on the input image. When the red-eye correction process performed in this way is larger than a predetermined correction level, a mark indicating that the red-eye correction process has been performed is added to the output image. On the other hand, the red-eye correction processing system 10 compares the information related to the eyes included in the input image with the information related to the eyes included in the image after the red-eye correction process. If the red-eye correction process performed in step S3 is smaller than a predetermined correction level, no mark or the like is added to the output image. In other words, the red-eye correction processing system 10 may perform the red-eye correction process even if the red-eye included in the image cannot be distinguished from the red-eye at first glance. In such a case, the red-eye correction processing system 10 indicates that the red-eye correction process has been performed. Do not add marks or the like to the output image. Further, the red-eye correction processing system 10 analyzes the tag information attached to the input image, and the degree of red-eye correction processing applied to the image is predetermined based on the imaging conditions when the image is captured. It may be determined whether or not the degree of correction is greater than. Then, the red-eye correction processing system 10 may add a predetermined mark or the like to the image based on the result determined from the analysis result of the tag information.

  Further, when the red-eye correction processing system 10 outputs the image after the red-eye correction processing, a mark or the like indicating information about the eyes included in the input image may be added to the output image. The red-eye correction processing system 10 adds a mark or the like corresponding to information for identifying a processing routine when the red-eye correction process is performed on the image to the image after the red-eye correction process is output. May be added. The processing routine is an example of a correction process that the red-eye correction processing system 10 performs on an image. For example, when the eye included in the image is a gold eye, the red-eye correction processing system 10 adds a mark or the like indicating that the correction processing routine applied to the gold eye is a gold eye correction process, and outputs the image. Good. The processing routine detects a red-eye correction processing routine when the red-eye correction processing is performed on the image, a gold-eye correction processing routine when the gold-eye correction processing is performed on the image, or an eye region that is an eye region from the image. It may be a face detection processing routine that detects a face area that sometimes includes eyes.

  Note that the red-eye correction processing system 10 may add a mark or the like to the image after the red-eye correction processing and output it as an index print, for example. When the red-eye correction processing system 10 has a display unit such as a monitor, an image with a mark or the like added to the display unit may be displayed. Further, the red-eye correction processing system 10 may record an image with a mark or the like on a recording medium such as a DVD. Further, the red-eye correction processing system 10 may print and output an image in which a mark or the like is added to a paper medium or the like. Note that an image to which a mark or the like is added may be output from any of the receiving machine 12, the image processing apparatus 14, the printer 16, and the personal computer 18.

  For example, the image data is acquired from the user in the receiving machine 12 installed in a store or the like. The accepting machine 12 that has acquired the image data may perform red-eye correction processing on the image in the accepting machine 12. Then, the accepting machine 12 may add a predetermined mark or the like to the image and display the image on the monitor of the accepting machine 12. The accepting machine 12 may transmit the acquired image data to the image processing apparatus 14 without performing the red-eye correction process on the acquired image. Then, the image processing device 14 that has received the image data may perform red-eye correction processing on the image. Subsequently, the image processing apparatus 14 may output an image to which a predetermined mark or the like is added. The image processing apparatus 14 may output image data by transmitting image data to which a predetermined mark or the like is added to the printer 16. When image data is input to the personal computer 18 owned by the user, the image processing device 14 acquires the image data input to the personal computer 18 via the network. Then, the image processing device 14 may perform red-eye correction processing on the acquired image.

  FIG. 2 shows an example of a functional configuration of the red-eye correction processing system 10 according to the present embodiment. The red-eye correction processing system 10 includes an image input unit 100, a red-eye detection unit 110, a black-eye detection unit 115, a red-eye correction unit 120, an imaging condition acquisition unit 130, a red-eye color information acquisition unit 135, a correction degree determination unit 140, and an imaging condition determination unit. 145, an image output unit 150, a correction result output unit 160, an index storage unit 170, a display unit 180, and a printing unit 185. In addition, when the red-eye correction processing system 10 is configured by a plurality of devices, each of the plurality of devices may include some or all of the above-described components. For example, when the printer 16 and the monitor are connected to the image processing device 14, the image processing device 14 includes the image input unit 100, the red eye detection unit 110, the black eye detection unit 115, the red eye correction unit 120, and the imaging condition acquisition unit. 130, a red-eye color information acquisition unit 135, a correction degree determination unit 140, and an imaging condition determination unit 145. The printer 16 includes an image output unit 150, a correction result output unit 160, an index storage unit 170, and a printing unit 185. The monitor may include a display unit 180.

  The image input unit 100 inputs a plurality of images. The image input unit 100 inputs an image from the outside as image data by reading a photograph and an image printed on a paper medium or the like. The image input unit 100 may input image data via a storage medium such as a semiconductor memory, or a network such as a LAN and the Internet. The image data includes image capturing information indicating imaging conditions when the image is captured, information indicating whether or not the image has been subjected to predetermined image processing, and an index print or image on any device. Tag information including information indicating whether or not to perform (for example, information compliant with Exif: Exchangeable Image File Format) may be added. The tag information may be added to the image in an imaging device that captures the image. The image input unit 100 supplies the input image data to the red-eye detection unit 110, the red-eye correction unit 120, and the black-eye detection unit 115. Further, the image input unit 100 supplies the tag information added to the image to the imaging condition acquisition unit 130 based on the control of the imaging condition acquisition unit 130.

  The red eye detection unit 110 detects red eyes included in each of the plurality of images received from the image input unit 100. Specifically, the red-eye detection unit 110 first receives a red-eye candidate having a red color included in a predetermined color range from the image input unit 100 using a predetermined detection threshold value for detecting red. Detect from multiple images. In such a case, the red color range may be set to a range in which red eyes included in the image can be appropriately detected. The red-eye detection unit 110 may set a red color range by machine learning. Further, the red-eye detection unit 110 may search for a red-eye candidate from the image by matching using a red-eye template image having a predetermined shape and size and having a color in a predetermined color range. Then, the red-eye detection unit 110 may have a red-eye template image having a plurality of different shapes and sizes in advance.

  Here, when searching for red eyes, the red eye detection unit 110 may search for red eyes based on at least one type of parameter. For example, the red-eye detection unit 110 uses a plurality of different-sized red-eye template images, a plurality of different-shaped red-eye template images, and a plurality of different types of hues as red-eye search conditions. A red-eye candidate may be detected using a red-eye template image. In addition, the red-eye detection unit 110 may move the red-eye template image on the image with a predetermined pitch width. The red-eye detection unit 110 has in advance information regarding a plurality of different pitch widths when moving the red-eye template image on the image, and a plurality of different movement start positions when moving the red-eye template image on the image. It may be. For example, the red eye detection unit 110 may detect a red eye candidate from the image by sequentially changing a plurality of parameters.

  Specifically, the red-eye detection unit 110 searches for red eyes in the image using the first parameter and tries to detect a red-eye candidate. If the red-eye candidate cannot be detected using the first parameter, the red-eye detection unit 110 attempts to detect the red-eye candidate by changing the first parameter to the second parameter. For example, the red-eye detection unit 110 detects a red-eye candidate by searching for a red-eye template image while moving the search position at the first pixel pitch in the image. If the red eye candidate cannot be detected at the first pixel pitch, the red eye detection unit 110 detects the red eye candidate while moving the red eye template image at a second pixel pitch different from the first pixel pitch. . The red-eye detection unit 110 may move the red-eye template image by sequentially changing the pixel pitch until a red-eye candidate is detected. Note that the red-eye detection unit 110 may use a pixel pitch at which the search position searched at the first pixel pitch and the search position searched at the second pixel pitch do not overlap.

  Then, the red-eye detection unit 110 selects a red-eye candidate at a predetermined position in the face among the detected red-eye candidates as a red-eye candidate that is a red-eye correction target. Specifically, the red eye detection unit 110 detects whether a face exists around the detected red eye candidate. For example, the red eye candidate corresponds to the eye in the face based on the positional relationship between the position of each part included in the face, such as the eyebrows, nose, mouth, ears, beard, and face contour, and the red eye candidate. It is determined whether or not it exists at the position. The red-eye detection unit 110 rotates the image of the subject existing around the detected red-eye candidate, and when the subject image has a symmetric shape, the subject image is a face image. You may judge that there is. Then, the red-eye detection unit 110 may select a red-eye candidate determined to be present at the position of the eye in the face as a red-eye candidate that is a red-eye correction target.

  Further, the red-eye detection unit 110 may match a red-eye candidate included in each of a plurality of images with a red-eye template image and detect a red-eye candidate having a predetermined first matching degree or more as a red eye. . The matching degree may be a matching degree with the red-eye candidate red-eye template image when the color, shape, and size of the red-eye template image are compared with the color, shape, and size of the red-eye candidate. . The matching degree may be higher as the matching degree between the red-eye template image and the red-eye candidate is higher. The higher the matching degree, the higher the probability that the red-eye candidate is an eye. Therefore, when the degree of matching is high, it may be evaluated that the eye reliability is high, which is an index indicating the probability that the red-eye candidate detected by the red-eye detection unit 110 is an eye. The red-eye detection unit 110 includes information indicating the size of the red-eye template image used for detecting the red-eye, information indicating the position occupied by the red-eye included in each of the plurality of images, information indicating the red-eye color, and the matching degree. Information indicating (or eye reliability) is supplied to the red-eye correction unit 120, the red-eye color information acquisition unit 135, and the correction degree determination unit 140. The information indicating the size of the red-eye template image is an example of red-eye area information.

  The black eye detection unit 115 detects black eyes included in each of the plurality of images received from the image input unit 100. Specifically, the black-eye detecting unit 115 first receives a plurality of black-eye candidates having black contained in a predetermined color range from the image input unit 100 using a predetermined detection threshold for detecting black. Are detected from the respective images. Then, by the same operation as the red-eye detection unit 110 in the above description, the black-eye detection unit 115 selects a black eye that is determined to exist at the position of the eye in the face from the detected black-eye candidates. The black eye detection unit 115 supplies information indicating the position occupied in the selected black eye face and information on the color of the black eye to the correction degree determination unit 140.

  The imaging condition acquisition unit 130 refers to the tag information added to the image input by the image input unit 100 and performs predetermined image processing on the image imaging information indicating the imaging condition when the image is captured. Information indicating whether or not an image is output, information indicating which index print or image is output in which device, and the like. The imaging condition acquisition unit 130 supplies the acquired information to the imaging condition determination unit 145. The imaging condition determination unit 145 analyzes the information acquired by the imaging condition acquisition unit 130 and determines the imaging condition of the image when the image input by the image input unit 100 is captured. The image capturing information included in the information analyzed by the image capturing condition determining unit 145 includes strobe information indicating whether or not the strobe is used at the time of image capturing, subject brightness information indicating the brightness of the subject included in the captured image, and an image. Subject distance information indicating the distance from the imaging device to the subject when the image is captured, light source information indicating the light source when the image is captured, imaging mode information indicating the type of imaging mode when the image is captured, and the input image May include red-eye correction information that is information indicating whether or not red-eye correction processing has been performed.

  Specifically, the imaging condition determination unit 145 determines whether or not the image input by the image input unit 100 was captured when the strobe light is emitted with reference to the strobe information. Further, the imaging condition determination unit 145 determines whether the luminance of the subject included in the image input by the image input unit 100 is equal to or higher than a predetermined value with reference to the subject luminance information. Note that the luminance of the subject may be the luminance of the background portion included in the image, the average luminance of the entire image, or the luminance distribution of the image. In addition, the imaging condition determination unit 145 determines whether the subject included in the image input by the image input unit 100 has a distance from the imaging device to the subject at the time of imaging within a predetermined range. Judge by referring to the distance information.

  Further, the imaging condition determination unit 145 determines whether an image is captured indoors from the light source information at the time of imaging of the image input by the image input unit 100 with reference to the light source information. For example, the imaging condition determination unit 145 determines whether the light source is sunlight, a fluorescent lamp, an incandescent lamp, or the like based on the wavelength distribution of light indicated by the light source information. That is, the imaging condition determination unit 145 analyzes the wavelength distribution of the light indicated by the light source information to determine whether the wavelength distribution is characteristic to sunlight or whether the wavelength distribution is characteristic to light emitted from a fluorescent lamp. The type of the light source may be determined by determining whether or not the wavelength distribution is characteristic to the light emitted by the incandescent lamp. In addition, the imaging condition determination unit 145 determines whether or not the image input by the image input unit 100 has been subjected to red-eye correction processing with reference to red-eye correction information. Note that the red-eye correction processing may be performed on the image in, for example, an image processing device or an image processing device different from the red-eye correction processing system 10.

  In addition, the imaging condition determination unit 145 is preliminarily set as a mode in which the imaging mode at the time of imaging of the image input by the image input unit 100 is a predetermined amount of light or less, or a mode in which a red eye phenomenon is highly likely to occur. Whether or not the mode is determined is determined with reference to the imaging mode information. For example, the imaging condition determination unit 145 determines whether the imaging mode at the time of imaging is a portrait mode or a night scene mode with reference to the imaging mode information. In addition, the imaging condition determination unit 145 is configured such that the imaging mode at the time of imaging is a party mode that is a mode for indoor party scenes, a night scene portrait mode that is a person imaging mode at night, or a mode for indoor imaging. Whether or not it is a certain indoor shooting mode may be determined by referring to the shooting mode information. Further, the imaging condition determination unit 145 may determine whether an image is captured in the red-eye reduction mode at the time of imaging. Note that the red-eye reduction mode may be an imaging mode that suppresses the occurrence of a red-eye phenomenon by a method of imaging a subject after pre-flashing the strobe. The imaging condition determination unit 145 supplies the determination result to the correction degree determination unit 140.

  The red-eye correction unit 120 performs red-eye correction processing that is included in the image received from the image input unit 100 and detected by the red-eye detection unit 110. The red-eye correction unit 120 detects a red eye to be subjected to red-eye correction in the image from information indicating a position in the image of the red-eye to be subjected to red-eye correction received from the red-eye detection unit 110, and performs a red-eye correction process on the detected red eye. Apply. Note that the red-eye correction process performed on the image by the red-eye correction unit 120 may include a red-eye area search process, a red-eye correction amount calculation process, and a red-eye black-eye conversion process. The red-eye area search process may be a process for specifying a red-eye area from red-eye detection information indicating a position or the like in the red-eye image detected by the red-eye detection unit 110. The red-eye correction amount calculation processing may be processing for calculating lightness / saturation adjustment amounts and replacement color information for the red-eye region specified in the red-eye region search processing. The red-eye / black-eye conversion processing may be processing for converting red-eye to black-eye from the red-eye region specified in the red-eye region search processing and the correction amount calculated in the red-eye correction amount calculation processing.

  Further, the red-eye correction unit 120 may perform not only red-eye correction processing for red eyes but also gold-eye correction processing for gold eyes. The gold eye correction process may be a process of changing the lightness and saturation and correcting the gold eye to the black eye after changing the color balance of the gold eye region to a predetermined balance. The gold eye correction process may be a process of correcting the gold eye by replacing the color of the gold eye region with a predetermined black eye color. Then, the red-eye correction unit 120 supplies information regarding the position in the red-eye image subjected to the red-eye correction process and the eye color after the red-eye correction process to the red-eye color information acquisition unit 135 and the correction degree determination unit 140. The red-eye correction unit 120 supplies the image after the red-eye correction processing to the image output unit 150 and the correction result output unit 160.

  The red-eye color information acquisition unit 135 acquires pre-correction color information that is information indicating the red-eye color before red-eye correction detected by the red-eye detection unit 110. Specifically, the red-eye color information acquisition unit 135 refers to the information indicating the position in the red-eye image included in each of the plurality of images and the information indicating the red-eye color received from the red-eye detection unit 110 before correction. Get color information. For example, the red eye color information acquisition unit 135 extracts the red eye color for each of the plurality of images. Then, the red-eye color information acquisition unit 135 may calculate the value of the extracted red-eye color R component, G component, and B component and acquire it as pre-correction color information. In addition, when the red-eye correction unit 120 does not perform the red-eye correction process, the red-eye color information acquisition unit 135 displays the information of the eye color, for example, the information of the black eye color that has not been subjected to the red-eye correction process. You may get as.

  Also, the red-eye color information acquisition unit 135 acquires post-correction color information that is information indicating the color of the red eye after the red-eye correction unit 120 performs the red-eye correction process. The red-eye color information acquisition unit 135 refers to the information received from the red-eye correction unit 120 indicating the position in the eye image after the red-eye correction process included in each of the plurality of images and the information indicating the eye color. Get back color information. For example, the red-eye color information acquisition unit 135 extracts the eye color after the red-eye correction process for each of the plurality of images. Then, the red-eye color information acquisition unit 135 may calculate the R component, G component, and B component level values of the extracted eye color as corrected color information. Further, the red-eye color information acquisition unit 135 may acquire information indicating the color balance in the red-eye region before red-eye correction detected by the red-eye detection unit 110 as pre-correction color information. The color balance may be, for example, a balance between the levels of the R component, the G component, and the B component. The red-eye color information acquisition unit 135 supplies the acquired pre-correction color information and post-correction color information to the correction degree determination unit 140.

  The correction degree determination unit 140 determines whether or not the correction degree of the red-eye correction process by the red-eye correction unit 120 is greater than a predetermined correction degree. For example, the correction degree determination unit 140 determines the correction degree when the level of the R component of the red eye color indicated by the pre-correction color information is higher than a predetermined level compared to the levels of the G component and the B component. You may judge it to be large. The correction degree determination unit 140 compares the pre-correction color information received from the red-eye color information acquisition unit 135 with the post-correction color information. Then, the degree-of-correction determination unit 140 performs the red-eye correction process of the red-eye correction unit 120 when the amount of change from the color indicated by the color information before correction is within a predetermined range. It is determined that the correction level is smaller than the predetermined correction level. For example, the correction degree determination unit 140 compares the level of the R component of the color indicated by the color information before correction with the level of the R component indicated by the color information after correction. That is, the correction degree determination unit 140 calculates the amount of change from the level of the R component of the color indicated by the color information before correction to the level of the R component of the color indicated by the color information after correction. Then, the correction degree determination unit 140 may determine that the correction degree is smaller than the predetermined correction degree when the calculated change amount is included in the predetermined range. On the other hand, the correction degree determination unit 140 may determine that the correction degree is greater than the predetermined correction degree when the calculated change amount exceeds a predetermined value.

  Furthermore, when the color balance indicated by the pre-correction color information acquired by the red-eye color information acquisition unit 135 is a predetermined balance, the correction degree determination unit 140 has a correction degree that is higher than a predetermined correction degree. You may judge it to be small. For example, the correction degree determination unit 140 compares the R component level, the G component level, and the B component level with respect to the color balance indicated by the pre-correction color information, respectively, and the color balance with a predetermined color balance ( For example, when the ratio between the R component level and the G component and B component levels is a predetermined ratio), it is determined that the correction degree is smaller than a predetermined correction degree.

  When the correction degree determination unit 140 determines that the black eye detected by the black eye detection unit 115 exists at the same position as the red eye detected by the red eye detection unit 110, the correction degree by the red eye correction process of the red eye correction unit 120 is determined. You may judge that it is smaller than the predetermined correction degree. Specifically, the correction degree determination unit 140 occupies the position occupied by the information received from the black eye detection unit 115 in the face of the black eye and the position occupied by the information received from the red eye detection unit 110 in the face of the red eye. To determine whether or not the black eye detected by the black eye detection unit 115 exists at the same position as the red eye detected by the red eye detection unit 110. Then, the correction degree determination unit 140 may determine that the correction degree of the red-eye correction unit 120 is smaller than a predetermined correction degree when the position of the black eye matches the position of the red eye.

  Here, the red eye candidate color detected by the red eye detection unit 110 and the black eye candidate color detected by the black eye detection unit 115 may partially overlap. For example, even when an R component is included in the black eye, the black eye detection unit 115 detects a black eye candidate from the image using a predetermined black eye detection threshold based on values such as brightness, saturation, and luminance. To do. Therefore, the black eye candidates detected by the black eye detection unit 115 may include eyes that are determined as red eye candidates by the red eye detection unit 110. In such a case, the correction degree determination unit 140 may determine that the degree of red-eye correction by the red-eye correction unit 120 is small. This is because the red-eye correction unit 120 performs red-eye correction processing for red eyes in which black-eye color components are mixed, compared to red-eye correction processing in which red-eye correction components 120 are mixed with black-eye color components. This is because the amount of correction for correcting the color of the black eye, for example, the amount of change for changing the levels of the R component, the G component, and the B component is small.

  The correction degree determination unit 140 refers to the matching degree received from the red-eye detection unit 110. The correction degree determination unit 140 corrects the red-eye when the red-eye correction unit 120 performs red-eye correction processing that is larger than the predetermined first matching degree and smaller than the predetermined second matching degree. It is determined that the correction degree of the red-eye correction process by the unit 120 is smaller than the predetermined correction degree. Thereby, even when the red eye correction is performed on the red eye whose eye reliability is equal to or less than a predetermined reference value, the correction degree is smaller than the predetermined correction degree for the red eye subjected to the red eye correction. I can judge. Further, the correction degree determination unit 140 may determine whether the correction degree of the red-eye correction process by the red-eye correction unit 120 is greater than a predetermined correction degree based on the determination result of the imaging condition determination unit 145. For example, the correction degree determination unit 140 determines that the correction degree of the image determined to be captured when the strobe light is not emitted by the imaging condition determination unit 145 is smaller than a predetermined correction degree. This is because the red-eye phenomenon or the gold-eye phenomenon rarely occurs when the strobe is not emitting light.

  Further, the correction degree determination unit 140 may determine that the correction degree of the image for which the luminance of the subject is determined to be equal to or higher than a predetermined value by the imaging condition determination unit 145 is smaller than the predetermined correction degree. . This is because if the luminance of the subject is greater than or equal to a predetermined value, it is highly likely that the subject was imaged during the day, and if the subject was imaged during the day, a red-eye phenomenon occurred. It is because the probability of doing is low. In addition, the correction degree determination unit 140 performs a predetermined correction on the correction degree of the image determined by the imaging condition determination unit 145 when the distance from the imaging device to the subject is not within a predetermined range at the time of imaging. You may judge that it is smaller than the degree. This is because the size of the subject's eye image present at a position more than a predetermined distance from the imaging device is small, so even if the red-eye phenomenon occurs, it may be difficult to determine the red-eye part. Because there is.

  Further, the correction degree determination unit 140 may determine that the correction degree of the image determined based on the light source information that the subject is imaged indoors by the imaging condition determination unit 145 is larger than a predetermined correction degree. . This is because, when a subject is imaged indoors, a strobe is often emitted, so that there is a high possibility that a red-eye phenomenon will occur. In addition, the correction degree determination unit 140 is configured to detect an image for which the subject is determined to have been captured in any one of the portrait mode, the night view mode, the party mode, the night view portrait mode, and the indoor shooting mode by the imaging condition determination unit 145. It is determined that the correction degree is larger than a predetermined correction degree. This is because a red-eye phenomenon occurs because the flash is often emitted when shooting the subject in any of the shooting modes of night view mode, portrait mode, party mode, night view portrait mode, or indoor shooting mode. It is because there is a high possibility of doing. Further, the correction degree determination unit 140 may determine that the correction degree is smaller than a predetermined correction degree when an image is captured in the red-eye reduction mode. This is because the subject has already been imaged by a method of reducing red eyes when the image is captured.

  Furthermore, the correction degree determination unit 140 determines that the correction degree of the image determined by the imaging condition determination unit 145 to have been subjected to red-eye correction processing is smaller than a predetermined correction degree. For example, when the imaging condition determination unit 145 determines from the information acquired by the imaging condition acquisition unit 130 that red-eye correction processing has been performed on the image in the imaging device that captured the image, the red-eye correction unit 120 It may be determined that the correction degree of the red-eye correction process is smaller than a predetermined correction degree.

  The correction degree determination unit 140 determines the correction degree when the size of the red-eye template image used for red-eye detection is larger than a predetermined size in the information received from the red-eye detection unit 110. You may judge that it is larger than the predetermined correction degree. On the other hand, if the size of the red-eye template image used for red-eye detection in the information received from the red-eye detection unit 110 is smaller than a predetermined size, the correction degree determination unit 140 determines the correction degree. It may be determined that the degree of correction is smaller than a predetermined correction degree. Further, the correction degree determination unit 140 may determine whether the correction degree is larger than a predetermined correction degree based on the position occupied in the red-eye image detected by the red-eye detection unit 110. Specifically, the correction degree determination unit 140 determines the correction degree in advance when the position of the red-eye image detected by the red-eye detection unit 110 is within a predetermined range from the edge of the image. It may be determined that the degree of correction is smaller. The correction degree determination unit 140 determines the correction degree in advance when the position of the red-eye image detected by the red-eye detection unit 110 is located at a predetermined distance from the center of the image. It may be determined that the degree of correction is smaller. This is because when the red eye is located at the end of the image, there is a high probability that the red eye is not the red eye of the main subject.

  Further, the correction degree determination unit 140 may determine the imaging scene in which the image is captured based on information (for example, auto setup) acquired from the image analysis information of the image input by the image input unit 100. . The imaging scene may be, for example, a daytime scene, a landscape scene, a night scene, or the like. For example, the correction degree determination unit 140 may determine that the correction degree of the red-eye correction processing by the red-eye correction unit 120 is smaller than a predetermined correction degree for images determined as daytime scenes and landscape scenes. . This is because, in daytime scenes and landscape scenes, images are often taken without flashing, so the probability that a red-eye phenomenon will occur is low.

  The correction degree determination unit 140 determines whether the correction degree of the red-eye correction process is greater than a predetermined correction degree by combining at least two of the results of the red-eye correction process and the plurality of determination criteria in the above description. May be. The degree-of-correction determination unit 140 includes information indicating the determination result, information about the eyes included in the image before correction (for example, information about the type of eyes, the position of the face of the eyes, the color of the eyes, etc.), and after correction Information relating to the eyes included in the image (for example, information regarding the position of the face in the corrected eye, the color of the corrected eye, and the correction processing routine) is supplied to the correction result output unit 160.

  The image output unit 150 outputs an image that has not been subjected to red-eye correction processing for an image for which red-eye has not been detected by the red-eye detection unit 110. Further, the image output unit 150 outputs an image on which red-eye correction processing has been performed for an image whose red-eye is detected by the red-eye detection unit 110. The image output unit 150 supplies the image that has not been subjected to the red-eye correction process and the image that has been subjected to the red-eye correction process to the display unit 180 or the printing unit 185. The image output unit 150 may record an image that has not been subjected to the red-eye correction process and an image that has been subjected to the red-eye correction process by writing them on a recording medium such as a DVD.

  The index storage unit 170 stores marks, character strings, and the like to be added to the index sprint output by the correction result output unit 160 in association with identification information for identifying the processing routine of the correction process. For example, the index storage unit 170 stores a mark having a red-eye shape indicating that the red-eye correction processing has been performed on the image, and a character string such as RedEye, in association with identification information for identifying the red-eye correction processing. ing. In addition, the index storage unit 170 stores a mark having a golden eye shape indicating that the gold correction process has been performed on the image and a character string such as GoldEye in association with the identification information for identifying the gold correction process. ing. Note that the marks, character strings, and the like stored in the index storage unit 170 are not limited to the above example, and may be predetermined marks that can identify predetermined correction processing. The index storage unit 170 supplies a predetermined mark or a predetermined character string to the correction result output unit 160 based on the control of the correction result output unit 160.

  The correction result output unit 160 performs red-eye correction processing on an image that has been subjected to red-eye correction processing greater than a predetermined correction level indicated by the determination result received from the correction level determination unit 140 among the images output from the image output unit 150. Information indicating that it has been applied is output in association with the image. The correction result output unit 160 also includes information indicating that the red-eye correction processing has been performed on an image that has been subjected to red-eye correction processing that is smaller than the predetermined correction degree indicated by the determination result received from the correction degree determination unit 140. Does not output in association with the image. Specifically, the correction result output unit 160 adds red-eye to an image for which the correction degree determination unit 140 determines that the correction degree of the red-eye correction processing by the red-eye correction unit 120 is smaller than a predetermined correction degree. Even when correction processing has been performed, information indicating that red-eye correction processing has been performed is not output in association with each other. Further, the correction result output unit 160 may output information indicating the type of eyes included in the image before the red-eye correction processing is performed in association with the image. On the other hand, the correction result output unit 160 performs red-eye correction processing on an image for which the correction degree determination unit 140 determines that the correction degree of the red-eye correction result by the red-eye correction unit 120 is greater than a predetermined correction degree. Corresponding information indicating that it has been applied is output.

  The correction result output unit 160 also corrects the red eye when outputting the image when the black eye detection unit 115 detects the black eye at the same position as the red eye detected by the red eye detection unit 110 from the image. Information indicating that processing has been performed is output without being associated. This is because even if the red eye is detected by the red eye detection unit 110, the red eye color that can be determined as the black eye by the black eye detection unit 115 is a red eye color that is not determined as the black eye by the black eye detection unit 115. This is because the degree is small. Accordingly, even if such red-eye correction processing is performed, the user may not be able to determine whether or not the red-eye correction processing has been performed. In such a case, the correction result output unit 160 outputs a mark or a character string indicating that the red-eye correction processing has been performed without associating the image with the image even if the image has been subjected to the red-eye correction processing.

  The correction result output unit 160 performs the correction process performed on the image that the correction degree determination unit 140 determines that the correction degree of the red-eye correction result by the red-eye correction unit 120 is larger than a predetermined correction degree. The mark or character string stored in the index storage unit 170 is extracted in association with the information to be identified. Then, the correction result output unit 160 outputs the extracted mark or character string in association with an image on which a predetermined correction process has been performed. The correction result output unit 160 supplies the image extracted from the index storage unit 170 or the image associated with the character string to at least one of the display unit 180 and the printing unit 185. The image that the correction result output unit 160 supplies to at least one of the display unit 180 and the printing unit 185 may be an index print image. That is, the correction result output unit 160 may output an image in which a predetermined mark or character string is associated and an image in which the predetermined mark and character string are not associated with each other in a predetermined layout.

  The display unit 180 displays the image received from the image output unit 150 on a display device such as a monitor. The display unit 180 is associated with the image received from the correction result output unit 160 and associated with information indicating that the red-eye correction process has been performed, and information indicating that the red-eye correction process has been performed. Display images that are not displayed. The printing unit 185 prints the image received from the image output unit 150 on a paper medium or the like. The printing unit 185 also associates the image received from the correction result output unit 160 with the information indicating that the red-eye correction process has been performed and the information indicating that the red-eye correction process has been performed. Print images that are not. For example, the printing unit 185 prints an image in the form of index printing.

  According to the red-eye correction processing system 10 according to the present embodiment, information indicating the correction process performed on the eye before the red-eye correction process can be recorded in association with the image. As a result, when an index print of the image is output, a mark or a character string indicating the correction processing applied to the image can be output together with the image. You can see at a glance that the correction process has been applied. The mark or character string indicating the correction process may be output adjacent to the image, or may be output superimposed on the image.

  In addition, according to the red-eye correction processing system 10 according to the present embodiment, the red-eye before red-eye correction is compared with the red-eye after red-eye correction, and red-eye correction is performed according to the degree of red-eye correction. Since it is determined whether or not a mark indicating that the image has been processed is associated with the image, a mark or character string indicating that the red-eye correction processing has been performed should not be output for an image that has been subjected to the minor correction processing. Can do. As a result, the red-eye correction process is performed, but if the correction result does not change much from the eye color of the original image, a mark or character string indicating that the red-eye correction process has been performed is not output. Therefore, it does not make the user feel dissatisfied.

  Furthermore, according to the red-eye correction processing system 10 according to the present embodiment, a mark or the like that explicitly indicates the content of the correction processing applied to the image can be output in association with the image. When charging is performed, it is possible to easily charge only the image that has been subjected to the correction process.

  FIG. 3 shows an example of the index print 200 according to the present embodiment. The correction result output unit 160, based on the determination result of the correction degree determination unit 140, indexes an image input by the image input unit 100 and an image obtained by performing red-eye correction processing on the image input by the image input unit 100. A predetermined mark can be added to the print for output. For example, consider a case where the image output unit 150 outputs a plurality of images 210, 212, 214, 216, 218, and the like to the index print 200. In such a case, the red-eye correction processing is performed on all of the image 210, the image 212, the image 214, the image 216, and the image 218, but the degree of the red-eye correction processing in the image 212 and the image 218 is higher than a predetermined degree. If smaller, the correction result output unit 160 outputs a mark 300, a mark 302, and a mark 304 indicating that red-eye correction processing has been performed on other images except the image 212 and the image 218.

  For example, a mark 300 indicating that the image 210 has been subjected to red-eye correction processing is added to the image 210. Even if the red-eye correction process is performed on the image 212, no mark is added to the image 212 if the correction degree of the red-eye correction process is smaller than a predetermined degree. Further, when the image 214 is captured indoors and the red eye correction process has been performed on the image 214, a mark 302 indicating that the red eye correction process has been performed is added. Furthermore, when the size of the eyes of the person included in the image 218 is smaller than a predetermined size, it is difficult to determine whether the eyes of the person included in the image 218 are red eyes. Regardless of whether or not the image 218 is subjected to the red-eye correction process, no mark is added to the image 218.

  According to the index print 200 according to the present embodiment, the user can grasp at a glance an image on which red-eye correction processing has been performed from among a plurality of images. In addition, it is possible to prevent a mark indicating that the red-eye correction process has been performed from being output for an image in which the degree of the red-eye correction process is smaller than a predetermined degree. As a result, a mark indicating that the red-eye correction processing has been performed is added to the image although the image has not been subjected to the appropriate red-eye correction processing even though the image has been subjected to appropriate red-eye correction processing. It is possible to prevent the user from being misunderstood if there is any.

  The correction result output unit 160 may add a frame, a mark, and the like corresponding to the eyes included in each of the plurality of input images to the output index print. For example, the correction result output unit 160 adds a red-eye mark or a red frame to the index print when the eyes included in the input image have red eyes. Further, the correction result output unit 160 adds a gold mark or a gold frame to the index print when the eyes included in the input image are gold. Further, the correction result output unit 160 does not add a mark or a frame to the index print when the eyes included in the input image are black eyes.

  Further, when displaying the image processing result on the display unit 180, the correction result output unit 160 includes a frame corresponding to an eye included in each of the plurality of input images, and a mark. Is displayed. For example, the correction result output unit 160 adds a red-eye mark or a red frame to the image displayed on the display unit 180 when the eyes included in the input image have red eyes. In addition, the correction result output unit 160 adds a gold mark or a gold frame to the image displayed on the display unit 180 when the eyes included in the input image are gold. Furthermore, the correction result output unit 160 does not add a mark or a frame to the image displayed on the display unit 180 when the eyes included in the input image are black eyes.

  Further, when the image is printed on a paper medium or the like and output from the printing unit 185, the correction result output unit 160 may add a predetermined frame or character string to the back surface of the output image. For example, the correction result output unit 160 adds a red-eye frame or a character string such as RedEye when the eyes included in the input image have red eyes on the back surface of the medium on which the image is printed. Further, the correction result output unit 160 adds a gold frame or a character string such as GoldEye to the back side of the medium on which the image is printed when the eyes included in the input image are gold. Further, the correction result output unit 160 does not add a frame and a character string when the eyes included in the input image are black eyes on the back surface of the medium on which the image is printed.

  This allows the user to determine at a glance the type of eye before the image is subjected to correction processing, and also to determine at a glance what type of correction processing has been applied to the image. Can do.

  FIG. 4 shows an example of the flow of processing in the red-eye correction processing system 10 according to the present embodiment. First, the red eye detection unit 110 detects red eyes from the image input by the image input unit 100 (S1000). The black eye detection unit 115 detects black eyes from the image input by the image input unit 100. The red eye detection unit 110 may include a plurality of eye detection units that detect eyes, and may include, for example, a red eye detection unit and a gold eye detection unit. The detection engine is an example of an eye detection unit. Then, the red eye detection unit 110 and the black eye detection unit 115 acquire information for identifying the type of the eye detection unit used for eye detection (S1010). The red eye detection unit 110 also acquires information indicating the reliability of the detected red eye or gold eye (S1010). Subsequently, the red-eye correction unit 120 performs a red-eye correction process on the red eye detected by the red-eye detection unit 110 (S1020).

  The red eye correction unit 120 may further acquire red eye correction information (S1030). The red-eye correction information is detected by the red-eye detection unit 110, for example, when the red-eye detected by the red-eye detection unit 110 is separated into the red-eye and the gold-eye in the gold-eye determination routine, and by the red-eye detection unit 110. It may be the reliability of red eyes. The gold eye determination routine may be executed by the red eye detection unit 110. For example, the gold eye determination routine determines whether the red eye candidate detected by the red eye detection unit 110 is a red eye or a gold eye respectively defined by a predetermined color range. Specifically, in the gold eye determination routine, the red eye detection unit 110 uses the red eye separation threshold having a higher R component level for at least one of the G component and B component levels than the detection threshold used by the red eye detection unit 110. It is determined whether the detected red eye candidate is a red eye or a gold eye.

  The gold eye determination routine also includes a first ratio indicating the ratio of the average value of the G component level to the sum of the average values of the R component, G component, and B component levels of the pixels included in the red eye candidate detected by the red eye detection unit 110. The second feature amount indicating the feature amount, the variance of the level of the R component of the image included in the red eye candidate detected by the red eye detection unit 110, or the average of the level of the R component of the pixel included in the red eye candidate detected by the red eye detection unit 110 Based on at least one feature amount of the third feature amount indicating the value, it may be determined whether the red eye candidate detected by the red eye detection unit 110 is a red eye or a gold eye. The red-eye correction unit 120 acquires the red-eye / gold-eye determination result in the gold-eye determination routine (S1030).

  Further, the red-eye correction unit 120 acquires the red-eye correction reliability (S1030). Note that the color indicated by the red eye before the red-eye correction unit 120 performs the red-eye correction process is included in a predetermined color range, and the red-eye shape is equal to or higher than a predetermined matching degree with the human eye shape. When matching is performed, it may be determined that the red-eye correction reliability is high. Subsequently, the correction degree determination unit 140 determines the type of eyes included in the original image (S1040). When the eyes included in the original image are eyes detected by the black eye detection unit 115, the correction degree determination unit 140 determines the eyes as black eyes and supplies the determination result to the correction result output unit 160. . Based on the received determination result, the correction result output unit 160 associates information indicating that the eyes included in the image are black eyes with the image when outputting the image (S1054).

  Further, the correction degree determination unit 140 determines whether the eye is an actual eye based on the eye reliability. For example, the correction degree determination unit 140 determines that the eye is an actual eye when the reliability of the eye received from the red-eye detection unit 110 exceeds a predetermined threshold. Moreover, the correction degree determination unit 140 determines that red-eye correction is appropriate when the reliability of red-eye correction exceeds a predetermined threshold. The correction result output unit 160 receives the eye reliability and the red eye correction reliability from the correction degree determination unit 140, and the received eye reliability and the red eye correction reliability exceed a predetermined threshold value, respectively. If not, it is determined that the eye has been detected by mistake or that the red-eye correction process has been applied to the image by mistake. Based on the determination by the correction degree determination unit 140, the correction result output unit 160 associates information indicating that the eye reliability included in the image is low with the image when outputting the image (S1056). . Alternatively, the correction result output unit 160 may not associate information indicating the eye reliability with the image when the reliability of the eye included in the image is smaller than a predetermined threshold.

  Further, when the red eye detection unit 110 detects a gold eye, the correction degree determination unit 140 supplies information indicating that the gold eye is detected to the correction result output unit 160. Then, when outputting the image, the correction result output unit 160 associates information indicating that the eye included in the image is a gold eye with the image (S1056). Further, the correction degree determination unit 140 determines that the eye included in the original image is not a gold eye or a black eye, and the reliability of the eye detected by the red eye detection unit 110 is high, and the red eye correction unit. If the reliability of red-eye correction at 120 is high, it is determined that the eye included in the image is red-eye. Then, based on the determination result of the correction degree determination unit 140, the correction result output unit 160 associates information indicating that the eyes included in the image are red eyes when the image is output (S1050). .

  FIG. 5 shows an example of a hardware configuration of the red-eye correction processing system 10 according to the present embodiment. The red-eye correction processing system 10 according to this embodiment includes a CPU peripheral unit having a CPU 1505, a RAM 1520, a graphic controller 1575, and a display device 1580 connected to each other by a host controller 1582, and a host controller by an input / output controller 1584. Input / output unit having communication interface 1530, hard disk drive 1540, and CD-ROM drive 1560 connected to 1582; Legacy having ROM 1510, flexible disk drive 1550, and input / output chip 1570 connected to input / output controller 1584 And an input / output unit.

  The host controller 1582 connects the RAM 1520 to the CPU 1505 and the graphic controller 1575 that access the RAM 1520 at a high transfer rate. The CPU 1505 operates based on programs stored in the ROM 1510 and the RAM 1520 to control each unit. The graphic controller 1575 acquires image data generated by the CPU 1505 or the like on a frame buffer provided in the RAM 1520 and displays it on the display device 1580. Alternatively, the graphic controller 1575 may include a frame buffer that stores image data generated by the CPU 1505 or the like.

  The input / output controller 1584 connects the host controller 1582 to the communication interface 1530, the hard disk drive 1540, and the CD-ROM drive 1560, which are relatively high-speed input / output devices. The communication interface 1530 communicates with other devices via a network. The hard disk drive 1540 stores programs and data used by the CPU 1505 in the red-eye correction processing system 10. The CD-ROM drive 1560 reads a program or data from the CD-ROM 1595 and provides it to the hard disk drive 1540 via the RAM 1520.

  The input / output controller 1584 is connected to the ROM 1510, the flexible disk drive 1550, and the relatively low-speed input / output device of the input / output chip 1570. The ROM 1510 stores a boot program executed when the red-eye correction processing system 10 is started, a program depending on the hardware of the red-eye correction processing system 10, and the like. The flexible disk drive 1550 reads a program or data from the flexible disk 1590 and provides it to the hard disk drive 1540 via the RAM 1520. The input / output chip 1570 connects various input / output devices via a flexible disk drive 1550, for example, a parallel port, a serial port, a keyboard port, a mouse port, and the like.

  The red-eye correction program provided to the hard disk drive 1540 via the RAM 1520 is stored in a recording medium such as the flexible disk 1590, the CD-ROM 1595, or an IC card and provided by the user. The red-eye correction program is read from the recording medium, installed in the hard disk drive 1540 in the red-eye correction processing system 10 via the RAM 1520, and executed by the CPU 1505. The red-eye correction program installed and executed in the red-eye correction processing system 10 operates on the CPU 1505 or the like to change the red-eye correction processing system 10 into the image input unit 100, the red-eye detection unit 110, and the black-eye detection described with reference to FIGS. 115, red-eye correction unit 120, imaging condition acquisition unit 130, red-eye color information acquisition unit 135, correction degree determination unit 140, imaging condition determination unit 145, image output unit 150, correction result output unit 160, index storage unit 170, display The unit 180 and the printing unit 185 function.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

1 is a diagram showing an outline of a red-eye correction processing system 10. FIG. 2 is a block diagram showing a functional configuration of a red-eye correction processing system 10. FIG. FIG. 3 is a diagram illustrating an index print 200. FIG. 4 is a flowchart showing a processing flow of the red-eye correction processing system 10. 2 is a block diagram illustrating a hardware configuration of a red-eye correction processing system 10. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Red eye correction processing system 12 Reception machine 14 Image processing apparatus 16 Printer 18 Personal computer 100 Image input part 110 Red eye detection part 115 Black eye detection part 120 Red eye correction part 130 Imaging condition acquisition part 135 Red eye color information acquisition part 140 Correction degree judgment part 145 Imaging condition determination unit 150 Image output unit 160 Correction result output unit 170 Index storage unit 180 Display unit 185 Printing unit 200 Index print 210, 212, 214, 216, 218 Image 300, 302, 304 Mark 1505 CPU
1510 ROM
1520 RAM
1530 Communication interface 1540 Hard disk drive 1550 Flexible disk drive 1560 CD-ROM drive 1570 Input / output chip 1575 Graphic controller 1580 Display device 1582 Host controller 1584 Input / output controller 1590 Flexible disk 1595 CD-ROM

Claims (16)

An image input unit for inputting a plurality of images;
A red-eye detection unit that detects red-eye included in each of the plurality of images input by the image input unit;
A red-eye correction unit for performing red-eye correction processing on the red eye detected by the red-eye detection unit;
And modifying degree determination section for determining whether greater or not than amendments degree of modification degree predetermined red-eye correction processing by the red eye correcting section,
An image that has not been subjected to red-eye correction processing is output for an image in which no red eye has been detected by the red-eye detection unit, and an image that has been subjected to red-eye correction processing is output for an image in which red eye has been detected by the red-eye detection unit An image output unit to
Among the images output by the image output unit, information indicating that the red-eye correction processing has been performed is output in association with the image for the image subjected to the red-eye correction processing greater than the predetermined correction degree. A red-eye correction processing system comprising: a correction result output unit that does not output information indicating that red-eye correction processing has been performed on an image that has been subjected to red-eye correction processing smaller than the predetermined correction degree; . A red-eye color information acquisition unit that acquires pre-correction color information that is information indicating a red-eye color before red-eye correction detected by the red-eye detection unit;
The correction degree determination unit, the front modified red color information acquiring unit has acquired on the basis of the color information, determines greater or not than amendments degree of modification degree predetermined red-eye correction processing by the red eye correcting section The red-eye correction processing system according to claim 1. The red-eye color information acquisition unit further acquires post-correction color information that is information indicating a red-eye color after the red-eye correction unit has performed red-eye correction processing,
The correction degree determination unit compares the pre-correction color information acquired by the red-eye color information acquisition unit with the post-correction color information, and determines the color indicated by the post-correction color information from the color indicated by the pre-correction color information. The red-eye correction processing system according to claim 2, wherein when the amount of change is included in a predetermined range, it is determined that the correction degree by the red-eye correction process of the red-eye correction unit is smaller than the predetermined correction degree. The red-eye color information acquisition unit acquires information indicating the color balance in the red-eye region before red-eye correction detected by the red-eye detection unit as pre-correction color information,
When the color balance indicated by the pre-correction color information is a predetermined balance, the correction degree determination unit determines that the correction degree by the red-eye correction process of the red-eye correction unit is smaller than the predetermined correction degree. The red-eye correction processing system according to claim 2. The red-eye detection unit further acquires red-eye area information indicating the size of the detected red-eye area,
The correction degree determination unit, according to claim wherein said red-eye detection unit based on the acquired eye region information to determine whether greater or not than amendments degree of modification degree predetermined red-eye correction processing by the red eye correcting section The red-eye correction processing system according to 1. The red-eye detection unit further acquires red-eye position information indicating a position in the detected red-eye image,
The correction degree determination unit, according to claim wherein said red-eye detection unit based on the acquired eye position information, determines whether larger or not than amendments degree of modification degree predetermined red-eye correction processing by the red eye correcting section The red-eye correction processing system according to 1. A black eye detection unit for detecting black eyes included in each of the plurality of images input by the image input unit;
When the black-eye detecting unit detects a black eye at the same position as the red eye detected by the red-eye detecting unit, the correction degree determining unit determines the correction degree determined by the red-eye correcting process of the red-eye correcting unit as the predetermined correction. The red-eye correction processing system according to claim 1, wherein the red-eye correction processing system is determined to be smaller than the degree. The red-eye detection unit matches a red-eye candidate and a red-eye template image included in each of the plurality of images, detects a red-eye candidate having a predetermined first matching degree or more as a red eye, and determines the correction degree When the red-eye correction unit has performed red-eye correction processing on red-eye smaller than a predetermined second matching level that is greater than the first matching level, the correction degree of the red-eye correction processing by the red-eye correction unit is The red-eye correction processing system according to claim 1, wherein it is determined that the correction degree is smaller than the predetermined correction degree. An image capturing condition determining unit that determines whether the image input by the image input unit is captured when the strobe is emitting light;
The correction level determination unit determines that the correction level of an image determined to be captured when no strobe light is emitted by the imaging condition determination unit is smaller than the predetermined correction level. The red-eye correction processing system described in 1. An imaging condition determination unit that determines whether or not the luminance of the subject included in the image input by the image input unit is greater than or equal to a predetermined value
Further comprising
The correction degree determination unit determines that the correction degree of the image for which the luminance of the subject is determined to be greater than or equal to a predetermined value by the imaging condition determination unit is smaller than the predetermined correction degree. The red-eye correction processing system according to claim 1 . An imaging condition determination unit that determines whether or not the distance from the imaging device to the subject at the time of imaging the image input by the image input unit is within a predetermined range
Further comprising
The correction degree determination unit determines that a correction degree for an image for which the distance is determined not to be within a predetermined range by the imaging condition determination unit is smaller than the predetermined correction degree. red-eye correction processing system according to 1. An imaging condition determination unit that determines whether or not the image is captured indoors from light source information at the time of imaging of the image input by the image input unit
Further comprising
2. The red-eye according to claim 1 , wherein the correction degree determination unit determines that a correction degree of an image determined by the imaging condition determination unit to be captured indoors is larger than the predetermined correction degree. Correction processing system. An imaging condition determination unit that determines whether an imaging mode at the time of imaging an image input by the image input unit is a portrait mode or a night view mode
Further comprising
The correction degree determination unit determines that a correction degree of an image for which the image is determined to be captured in a portrait mode or a night view mode by the imaging condition determination unit is larger than the predetermined correction degree. Item 2. The red-eye correction processing system according to Item 1 . An imaging condition determination unit that determines whether or not red-eye correction processing has been performed on the image input by the image input unit
Further comprising
The correction degree determination unit determines that a correction degree of an image determined by the imaging condition determination unit to be subjected to red-eye correction processing is smaller than the predetermined correction degree. red-eye correction processing system according to 1. A red-eye correction processing method for performing red-eye correction processing on an image,
An image input stage for inputting a plurality of images;
A red-eye detection step of detecting red-eye included in each of the plurality of images input in the image input step;
A red-eye correction step of performing red-eye correction processing on the red eye detected in the red-eye detection step;
A modified degree determination step of modifying the degree of red eye correcting process by the red eye correcting step determines greater or not than amendments predetermined degree,
An image that has not been subjected to red-eye correction processing is output for an image in which red-eye is not detected in the red-eye detection step, and an image that has been subjected to red-eye correction processing is output for an image in which red-eye is detected in the red-eye detection step An image output stage,
Among the images output in the image output stage, information indicating that the red-eye correction processing has been performed is output in association with the image for the image subjected to the red-eye correction processing greater than the predetermined correction degree. A correction result output step that does not output information indicating that the red-eye correction processing has been performed on the image that has been subjected to the red-eye correction processing smaller than the predetermined correction degree in association with the image. Method. A red-eye correction processing program for a red-eye correction processing system that performs red-eye correction processing on an image, the red-eye correction processing system comprising:
An image input unit for inputting a plurality of images,
A red-eye detection unit that detects red-eye included in each of the plurality of images input by the image input unit;
A red-eye correction unit for performing red-eye correction processing on the red eye detected by the red-eye detection unit;
Fixed degree determining section for determining whether greater or not than amendments degree of modification degree predetermined red-eye correction processing by the red eye correcting section,
An image that has not been subjected to red-eye correction processing is output for an image in which red-eye is not detected by the red-eye detection unit, and an image that has been subjected to red-eye correction processing is output for an image in which red-eye is detected by the red-eye detection unit Image output unit,
Among the images output by the image output unit, information indicating that the red-eye correction processing has been performed is output in association with the image for the image subjected to the red-eye correction processing greater than the predetermined correction degree. A red-eye correction processing program that functions as a correction result output unit that does not output information indicating that red-eye correction processing has been applied to an image that has been subjected to red-eye correction processing smaller than the predetermined correction degree. .

Images